t5-small-summarize-billsum

This model is a fine-tuned version of google-t5/t5-small on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.3984
• Rouge1: 0.2025
• Rouge2: 0.0983
• Rougel: 0.1707
• Rougelsum: 0.1707
• Gen Len: 20.0

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 1e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	124	2.8492	0.13	0.0388	0.1087	0.1089	20.0
No log	2.0	248	2.6212	0.1398	0.0499	0.1155	0.1152	20.0
No log	3.0	372	2.5327	0.148	0.0567	0.1237	0.1235	20.0
No log	4.0	496	2.4869	0.1627	0.0672	0.1345	0.1343	20.0
3.0124	5.0	620	2.4531	0.1842	0.0819	0.152	0.1518	20.0
3.0124	6.0	744	2.4297	0.194	0.088	0.1616	0.1617	20.0
3.0124	7.0	868	2.4147	0.1981	0.0934	0.1666	0.1665	20.0
3.0124	8.0	992	2.4061	0.2022	0.098	0.1705	0.1705	20.0
2.6076	9.0	1116	2.4001	0.2025	0.0983	0.1707	0.1707	20.0
2.6076	10.0	1240	2.3984	0.2025	0.0983	0.1707	0.1707	20.0

Framework versions

• Transformers 4.55.2
• Pytorch 2.8.0+cu128
• Datasets 3.6.0
• Tokenizers 0.21.4